Runaway argument? [closed]

Question

Suddenly my document stopped producing pdf and I can't seem to find out what's causing this error:

Runaway argument? \let \AtEndDocument \@firstofone \@enddocumenthook \@checkend {docume\ETC. ! File ended while scanning use of \beamer@collect@@body. \par <*> example.tex

?
Process interrupted by user

This is what I get afterwards:

And deleting those files didn't fix the issue:

What should I do? I'm not sure if the post should include my code since it's long and I don't know which part is causing error to paste it here.

Here's the MWE:

\documentclass{beamer}
\usepackage{graphicx}

\usetheme{Warsaw}
 \title[Artificial Neural Network]{Introduction  to ANN\\Artificial Neural Network}
 \author{Gigili}
 \date{January 6, 2013}
  \begin{document}

  \begin{frame}
  \titlepage{}

  \end{frame}

 \begin{frame}
 We are drowning in information and starving for knowledge.
  \hspace*{70mm}      – Rutherford D. Roger

  \end{frame}



  \begin{frame}{definitions of ANN}

 \centerline{\includegraphics[scale=0.6]{anatomy}}

  end{frame}

   \begin{frame}{Important factors to consider choosing an ANN architecture}

  - the number and type of neuronal layers and the number of neurons comprising each layer

  - Creating the NN architecture therefore means coming up with values for the number of 
            layers of each type and the number of nodes in each of these layers.

      Following this schema this will give you a competent architecture but probably not an
       optimal one.

     But once this network is initialized, you can iteratively tune the configuration 
        during training using a number of ancillary algorithms; one family of these works
       by pruning nodes based on (small) values of the weight vector after a certain number
          of training epochs--in other words, eliminating unnecessary/redundant nodes.



        \end{frame}



         \begin{frame}{main types of neural network architecture}

      - Feed Forward such as MLP: capable of approximating continuous and integrable 
  functions, consists of one input and one output layer and one or more hidden layers.

  The activities of the neurons in each layer are a non-linear function of the activities
  in the layer below.


      - Recurrent: These have directed cycles in their connection graph. 
       That means you can sometimes get back to where you started by following the arrows.


           - Symmetrically connected networks: These are like recurrent networks, 
            but the connections between units are symmetrical 
             (they have the same weight in both directions).



       \end{frame}

       \begin{frame}

         \centerline{\includegraphics[scale=0.6]{MLP}}

         \end{frame}

        \begin{frame}

         \centerline{\includegraphics[scale=0.6]{recurrent}}
       \begin{center}Recurrent structure \end{center}


              \end{frame}



   \begin{frame}{classfication using NN- continue}

      Possible solutions:

      1) Trial and error (experience)\\
       2) adds and delets neurons as needed during training \\
        3) constructive algorithms, network pruning, regularization

       \end{frame}
      \end{document}